Research map / Video / Video generation
Audio2Gestures: Generating Diverse Gestures From Speech Audio With Conditional Variational Autoencoders
Jing Li, Di Kang, Wenjie Pei, Xuefei Zhe, Ying Zhang, Zhenyu He, Linchao Bao
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add split shared/motion-specific latent code with relaxed motion loss (ablation) | baseline · Trinity (3D) | L1 joint error: 8.22 -> 8.69 | Hurt |
| add split latent code (ablation) | baseline · Trinity (3D) | PCK: 0.80 -> 0.77 | Hurt |
| add split latent code (ablation) | baseline · Trinity (3D) | Multimodality: 5.90 (baseline has none) | Improved |
| add mapping network (ablation) | + split · Trinity (3D) | L1 joint error: 8.69 -> 8.06 | Improved |
| add mapping network (ablation) | + split · Trinity (3D) | PCK: 0.77 -> 0.80 | Improved |
| add mapping network (ablation) | + split · Trinity (3D) | Multimodality: 5.90 -> 3.44 | Hurt |
| add bicycle constraint (ablation) | + mapping net · Trinity (3D) | Multimodality: 3.44 -> 3.68 | Improved |
| add bicycle constraint (ablation) | + mapping net · Trinity (3D) | Diversity: 5.86 -> 6.31 | Improved |
| add bicycle constraint (ablation) | + mapping net · Trinity (3D) | L1 joint error: 8.06 -> 7.94 | Improved |
| add diversity loss (ablation) | + bicycle constraint · Trinity (3D) | Multimodality: 3.68 -> 4.11 | Improved |
| add diversity loss (ablation) | + bicycle constraint · Trinity (3D) | L1 joint error: 7.94 -> 7.84 | Improved |
| add diversity loss (ablation) | + bicycle constraint · Trinity (3D) | PCK: 0.80 -> 0.82 | Improved |
Closest papers by meaning
- Audio-Driven Co-Speech Gesture Video Generation (NeurIPS 2022 · similarity 0.90)
- Co-Speech Gesture Video Generation with Implicit Motion-Audio Entanglement (CVPR 2025 · similarity 0.89)
- Towards Variable and Coordinated Holistic Co-Speech Motion Generation (CVPR 2024 · similarity 0.89)
- ConvoFusion: Multi-Modal Conversational Diffusion for Co-Speech Gesture Synthesis (CVPR 2024 · similarity 0.89)
- TANGO: Co-Speech Gesture Video Reenactment with Hierarchical Audio Motion Embedding and Diffusion Interpolation (ICLR 2025 · similarity 0.88)
- Co-Speech Gesture Video Generation via Motion-Decoupled Diffusion Model (CVPR 2024 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
